| Part Ⅰ Association of Lipid Levels with Ischemic and Hemorrhagic Stroke:A Prospective Cohort Study among 267,500 ChineseBackground and objectiveStroke continues to be one of the greatest causes of deaths,with total deaths increasing from 4.6 million in 1990 to 6.4 million in 2013 worldwide.In China,stroke has become the leading cause of death in recent years.Data from National Disease Surveillance Points System suggested that nearly 2.4 million new strokes and 1.1 million stroke-related deaths were identified annually.The associations between lipids and stroke were controversial in previous studies.The association between lipids and stroke especially type-specific stroke needs to be further investigated for the lack of large,prospective studies among Chinese population.We investigated the association of lipids with total,ischemic,and hemorrhagic stroke among Chinese population.Subjects and methodsParticipants for the current study were from 6 studies,the Asia Pacific Cohort Studies Collaboration(APCSC),China Multicenter Collaborative Study of Cardiovascular Epidemiology(ChinaMUCA)1992-1994,ChinaMUCA 1998,International Collaborative Study of Cardiovascular Disease in Asia(InterASIA),Kailuan Study,and Community Intervention of Metabolic Syndrome in China&Chinese Family Health Study(CIMIC).A total of 267,500 study participants without coronary heart disease and stroke at baseline were prospectively investigated.In each cohort from APCSC,age,sex,smoking status,and alcohol consumption were recorded whenever available.In the other 5 studies,a standardized and validated questionnaire administered by well-trained staff or research doctors was employed to collect information on socio-demographic characteristics,personal medical history,and lifestyle risk factors.Overnight fasting venous blood samples were drawn by venipuncture to measure lipid levels.Disease status was ascertained with stringent standard at the end point of each cohort.Cox proportional hazards regression models were used to estimate the hazard ratios(HRs)and their corresponding 95%confidence intervals(CIs)of lipid levels with total,ischemic,and hemorrhagic stroke events with adjustment for multiple variables.Multivariate-adjusted restricted cubic spline analyses were used to describe the relationship of lipid levels and total,ischemic,and hemorrhagic stroke events.ResultsOver a follow-up of 2,295,881 person-years,8,072 stroke events(5,458 ischemic stroke,2,186 hemorrhagic stroke,and 603 unspecified stroke events)were identified.The longest median follow-up duration of the cohorts was 19 years.After adjustment for age,gender,geographic region(north vs south),smoking status,alcohol consumption,education level,hypertension,and body mass index,an increase of 38.6 mg/dl(1 mmol/L)in total cholesterol(TC)level corresponded to 4%and 8%higher risk of total and ischemic stroke,respectively.An increase of 38.6 mg/dl(1 mmol/L)in low-density lipoprotein cholesterol(LDL-C)level corresponded to 5%and 8%higher risk of total and ischemic stroke,respectively.An increase of 88.5 mg/dl(1 mmol/L)in triglyceride(TG)level corresponded to 6%and 7%higher risk of total and ischemic stroke,respectively.Compared with participants with high-density lipoprotein cholesterol(HDL-C)level 50-59.9 mg/dl,HRs(95%CIs)were 1.28(1.18-1.38),1.15(1.08-1.23)for total stroke,and 1.23(1.12-1.35),1.13(1.04-1.22)for ischemic stroke in participants with HDL-C<40 and 40-49.9 mg/dl,respectively.Compared with participants with TC level 160-199.9 mg/dl,HRs(95%CIs)were 1.43(1.11-1.85)for hemorrhagic stroke in participants with TC<120 mg/dl.Compared with participants with HDL-C level 50-59.9 mg/dl,HRs(95%CIs)were 1.28(1.10-1.49)and 1.17(1.03-1.33)for hemorrhagic stroke in participants with HDL-C<40 and 40-49.9 mg/dl,respectively.Multivariate-adjusted restricted cubic spline analyses showed a linear relationship of TC,LDL-C with total stroke(both P<0.001),ischemic stroke(both P<0.001),and linear relationship of TC with hemorrhagic stroke(P = 0.029).Non-linear relationship of HDL-C,TG with total stroke(both P<0.001)and ischemic stroke(both P<0.001),and linear relationship of HDL-C with hemorrhagic stroke(P<0.001)were identified.ConclusionThe study identified a positive association of TC,LDL-C,and TG with total and ischemic stroke.The risk of total and ischemic stroke increased among participants with HDL-C less than 50 mg/dl.The risk of hemorrhagic stroke increased among participants with TC less than 120 mg/dl or HDL-C less than 50 mg/dl.Level of TC,LDL-C,and TG should be controlled as low as possible with HDL-C controlled>50 mg/dl.The relationship between lipid level and subtype of stroke provided more evidence to make specific intervention on lipids for the prevention of stroke.Part Ⅱ Usefulness of Non-High-Density Lipoprotein Cholesterol as Risk Factor of Cardiovascular Disease in ChineseBackground and objectiveCardiovascular diseases(CVDs)contribute to the greatest proportion of deaths related to noncommunicable diseases.All ages death attributable to CVD has increased from 12.28 million in 1990 to 17.30 million in 2013 globally.Global cardiovascular disease deaths were projected to increase to 23.3 million in 2030.According to the data from Global Burden of Disease Study 2013,all ages death attributable to CVD has increased from 2.56 million in 1990 to 3.72 million in 2013 in China.The National Lipid Association(NLA)and National Institute for Health and Care Excellence(NICE)announced that non-high-density lipoprotein cholesterol(non-HDL-C)was a better target than low-density lipoprotein cholesterol(LDL-C)for primary prevention of CVD.The predictive effect of non-HDL-C for CVD in Chinese general population has not been well demonstrated.We examined the relation between non-HDL-C and CVD and compare the effect of non-HDL-C and LDL-C on CVD risk in Chinese population.Subjects and MethodsBaseline examination of participants aged 35 to 74 years from the China Multicenter Collaborative Study of Cardiovascular Epidemiology(ChinaMUCA)1998 and the International Collaborative Study of Cardiovascular Disease in Asia(InterASIA)was conducted in 1998-2001.A standard questionnaire was administered to obtain information on demographic characteristics and lifestyle risk factors by trained staff.Participants were invited to take a physical examination after the interview.Overnight fasting blood samples were drawn by venipuncture to measure lipid and glucose levels.Follow-up evaluation was conducted among 19,268 participants in 2007-2008.Cox proportional hazards regression models were used to obtain the multivariable-adjusted hazard ratios(HRs)and 95%confidence intervals(CIs)for CVD,coronary heart disease(CHD),and stroke with non-HDL-C<130 mg/dl as the reference group.In addition,we estimated HRs of per-unit(30 mg/dl)increase of non-HDL-C or LDL-C level for overall population and among subgroups for hypertension,diabetes,body mass index,and triglyceride(TG).The likelihood ratio test was used to compare the effects of non-HDL-C and LDL-C on CVD,CHD,and stroke risk.ResultsA total of 656 CVD events,224 CHD events and 387 stroke events were identified during a mean follow up of 7.9 years(152 246 person-years).The mean follow-up duration of ChinaMUCA 1998 and InterASIA were 9.0 years and 7.0 years,respecitvely.Compared with those with non-HDL-C level of<130 mg/dl,multivariable adjusted HRs(95%CI)of CVD were 1.12(0.92-1.36),1.30(1.04-1.62)and 1.93(1.50-2.47)in participants with non-HDL-C levels of 130-159.9 mg/dl,160-189,9 mg/dl and>190 mg/dl,respectively.Multivariable adjusted HRs of CHD were 1.59(1.14-2.22),1.85(1.27-2.71),and 3.34(2.25-4.96)in participants with non-HDL-C levels of 130-159.9 mg/dl,160-189.9 mg/dl and>190 mg/dl,respectively.Multivariable adjusted HRs of stroke were 1.12(0.87-1.43),1.46(1.11-1.92),and 1.80(1.30-2.49)in participants with non-HDL-C levels of 130-159.9 mg/dl,160-189.9 mg/dl and>190 mg/dl,respectively.An increase of 30 mg/dl in non-HDL-C level would correspond to 15%,24%,and 12%increase in risk of CVD,CHD,and stroke,respectively.Using likelihood ratio tests,non-HDL-C appeared to have similar effect on CVD risk(χ2 for non-HDL-C,18.01,P<0.001;χ2 for LDL-C,18.90,P<0.001),CHD risk(χ2 for non-HDL-C,15.58,P<0.001;χ2 for LDL-C,16.49,P<0.001),and stroke risk(χ2 for non-HDL-C,6.91,P = 0.009;χ2 for LDL-C,7.39,P<0.007)as LDL-C.The equal effect of non-HDL-C and LDL-C on CVD,CHD,and stroke was also persistent in subgroup analyses according to hypertension,diabetes,body mass index,and TG.ConclusionIn conclusion,higher non-HDL-C level is associated with the increased risk of CVD,coronary heart disease,and stroke.Non-HDL-C has a similar effect as LDL-C on CVD,coronary heart disease,and stroke risk in Chinese.The result from our study provided more evidence on the management of lipid levels and primary prevention of CVD.Furthermore,non-HDL-C could be considered as the screening indicator for primary prevention of CVD.Part Ⅲ Metabolites Quantified by Untargeted Metabolomics Associated with Total and Lipoprotein Cholesterol LevelBackground and objectiveCholesterol is the key constituent in cell membranes and precursor of bile acids and steroid hormones.Total cholesterol(TC),its two main fractions low-density lipoprotein cholesterol(LDL-C)and high-density lipoprotein cholesterol(HDL-C),and triglyceride(TG)have been identified as the risk factors for coronary heart disease.In the development of disease,progress related to total and lipoprotein cholesterol levels such as atherosclerosis may have started early in life before diagnosed as dyslipidemia.Only focusing on the clinical measurement of total and lipoprotein cholesterol levels would not allow prevention of those metabolic abnormalities and irreversible pathology that occurred in early stage.We investigated the association of metabolites and serum metabolite modules with total and lipoprotein cholesterol levels.Subjects and MethodsThe Bogalusa Heart Study(BHS)is a community-based long-term study investigating the natural history of cardiovascular disease among white and African-American.Data generated from 2013 to 2016 follow-up visit were leveraged in the current metabolomics study of these participants.The BHS metabolomics study finally included in the current analysis.Questionnaires were administered at clinic visit to obtain information on demographic characteristics,lifestyle risk factors,and personal medical history.Body weight and height were measured twice to the nearest 0.1 kg and 0.1 cm,respectively.Participants were instructed to fast for 12 hours before blood sample collection.Serum TC,HDL-C and TG levels were assayed using an enzymatic procedure as part of a lipid panel and LDL-C was calculated using the Friedewald equation for those with TG less than 400 mg/dl.Detection and quantification of metabolites were completed by Metabolon Inc(Durham,NC)using an untargeted,ultrahigh performance liquid chromatography-tandem mass spectroscopy(UPLC-MS/MS).The multiple linear regression model was used to analyze the association of each metabolite with total and lipoprotein cholesterol levels.The race-stratified analysis was performed to identify those metabolites that were significantly associated and had concordant direction with total and lipoprotein cholesterols in both white and African-American.To identify modules of co-abundant serum metabolites for white and African-American,weighted correlation network analysis(WGCNA)was used.The correlation between each serum metabolite module and covariates adjusted total and lipoprotein cholesterol levels was further examined.Bonferroni correction was used to adjust for multiple test.ResultsAfter multivariate adjustment,126,79,36,and 136 metabolites were associated with TC,LDL-C,HDL-C,and TG and had concordant direction of association in white and African-American after adjusting for multiple test,respectively.In total,206 unique metabolites were identified to be associated with any lipid level across different race groups.These metabolites were from 7 super-pathways,with 169 from lipid pathway,8 from amino acid pathway,5 from cofactors and vitamins pathway,2 from carbohydrate pathway,2 from peptide pathway,1 from nucleotide pathway,and 19 from unknown pathway.Among those unique metabolites associated with any lipid level across race groups after Bonferroni correction,11,2,8,and 4 metabolites with named compounds were the novel ones that were neither previously reported nor located in the sub-pathway previously reported to be associated with TC,LDL-C,HDL-C,and TG,respectively.For example,TC,LDL-C,and HDL-C respectively increased 9.60,5.92,and 3.04 mg/dl among white(P = 1.70×10-11 3.74×10-6,and 2.74×10-9,respectively)and 15.41,10.32,and 4.15 mg/dl among African-American(P = 1.90×10-12,4.89×10-7,and 5.40×10-7,respectively)with per SD increase of 2-hydroxypalmitate from fatty acid,monohydroxy sub-pathway.WGCNA identified 15 and 6 serum metabolite modules for white and African-American,respectively.Of these metabolite modules,10 and 5 associated with at least 1 lipid trait after adjustment for multiple test among white and African-American,respectively.For example,module 2 identified among White which comprised diacylglycerol,monoacylglycerol,ceramides,and sphingolipid metabolism was positively correlated with TC(correlation coefficient = 0.48,P = 1 × 10-49),LDL-C(correlation coefficient = 0.28,P = 2 × 10-16),and TG(correlation coefficient = 0.77,P =2 × 10-160),and negatively correlated with HDL-C(correlation coefficient =-0.23,P = 4× 10-11).ConclusionThis study reported the novel association of single metabolites,serum metabolite modules with total and lipoprotein cholesterol and provided clues for early prevention of metabolic abnormalities and irreversible pathology that may result from dyslipidemia.Large and prospective studies are warranted to replicate the findings from our study. |
PDF Full Text Request